Dimension P - New PT2399 Chorus

[CynicalMan]

So, I got a Tayda order about a week ago, including some new PT2399s. Excited about my new toys, I decided to design a two voice chorus à la Boss Dimension C. Having heard on this forum about how nice it sounded compared to regular choruses (mostly from Mark Hammer  ) , I wanted to try multiphase chorusing out for myself. It turned out well IMO, a lot lusher than single voice choruses. It's a bit too much for distortion, but the clean sounds are great.

Schematic (click for bigger image)


I know it looks big, but it's really not that complicated for a chorus. It has 55 parts, compared to 48 for the Little Angel and 68 for the Small Clone. Granted, it's probably has more noise than they do, but I'm working on that.

This chorus works like the Boss DC-2, with two delay lines, one with its LFO signal inverted. This means that while one is pitch shifting up the other is pitch shifting down, and vice versa. Each of these signals is then mixed together with the dry signal. Note that the delay resistors are different, one at 270 ohms and the other at 100 ohms. This is because I was getting TZF with them at the same value, and it was making it sound too wobbly. The delay times are still close enough to get a bit of the flanger flavour as they approach each other. I've tried to keep the design as simple as possible, but there's room for further improvement and mods. I'm probably going to throw in a 1/2 Voice switch on mine, and I want to try a quadrature oscillator instead of the inverted LFO. With some clever switching, you could get a stereo/mono switch in there, with one voice and the dry signal going to each channel.

I owe a big debt to frequencycentral and anchovie for this, it's heavily influenced by both the Little Angel and the One Chip Chorus.

The circuit definitely isn't finalized yet. I'm still tweaking it, and I'm looking forward to suggestions from you guys.

Clips coming soon.

[electricteeth]

Neat Neat! Looks really cool. I just ordered a bunch of those pt's in the mail! Clips?

[WhenBoredomPeaks]

I was getting TZF

Whoa, i want this. I gotta order some PT2399s. I remember 2 years ago when i made my Rebote 2.5 i disliked what it did to the repeats but nowadays i am into lo-fi and sounding bad so i actually prefer it to clean digital delays. I can't wait to hear the chip in a chorus though i don't think that it will really gonna color the sound of the repeats because the delay time is really low.

[frequencycentral]

Very cool! But why avoid the TZF? Why not slow the LFO down and make a feature of it?

[WhenBoredomPeaks]

Very cool! But why avoid the TZF? Why not slow the LFO down and make a feature of it?

I am pretty sure that if i gonna build this i will use a 1k pot instead of one of the delay resistors. So during the pots sweep i could get TZF and bigger or smaller differences between the two stages than the stock build for possible sonic exploration.

Btw there was a topic a while ago that we should ground the delay resistor (coming from pin6) into pin4 (for lower noise) but pin 4 (digital ground) is unused here, why?

edit: here is the "digital ground and PT2399" thread: http://www.diystompboxes.com/smfforum/index.php?topic=92528.0

[Slade]

Very nice! Can't wait to hear it.

[DougH]

Very cool! But why avoid the TZF? Why not slow the LFO down and make a feature of it?

I am pretty sure that if i gonna build this i will use a 1k pot instead of one of the delay resistors. So during the pots sweep i could get TZF and bigger or smaller differences between the two stages than the stock build for possible sonic exploration.

There is delay control like that on the Liquiflange. When set to zero you get a nice TZF sound.

I'm really impressed with the advanced projects that have been coming out of this forum the last few years- especially the "breakthroughs" that have made some of the more complex circuits easier to build for diy-ers. Very cool.;-)

[CynicalMan]

Neat Neat! Looks really cool. I just ordered a bunch of those pt's in the mail! Clips?

On their way.

I was getting TZF

Whoa, i want this. I gotta order some PT2399s. I remember 2 years ago when i made my Rebote 2.5 i disliked what it did to the repeats but nowadays i am into lo-fi and sounding bad so i actually prefer it to clean digital delays. I can't wait to hear the chip in a chorus though i don't think that it will really gonna color the sound of the repeats because the delay time is really low.

There is filtering, but no audible digital distortion. So it's not the same kind of colouring that you'd get with a rebote.

Very cool! But why avoid the TZF? Why not slow the LFO down and make a feature of it?

A few reasons. It's not really the sound I wanted for this, and it makes the LFO sound more like a full-wave rectified sine wave than a triangle wave. The flanging sound isn't that strong because there's no regeneration, it just sounds like a chorus with a throbby LFO. Dialing it in is tricky too, you have to use a trimpot or the zero crossings won't be in the middle of the LFO wave. Finally, I got a bit of a whining sound that occurred at the zero crossings, which I think is beating from the clock frequencies being so close together.

I'd like to have a go at a PT2399 flanger, but I don't know how one could manage regeneration with the 30ms latency.

I am pretty sure that if i gonna build this i will use a 1k pot instead of one of the delay resistors. So during the pots sweep i could get TZF and bigger or smaller differences between the two stages than the stock build for possible sonic exploration.

Btw there was a topic a while ago that we should ground the delay resistor (coming from pin6) into pin4 (for lower noise) but pin 4 (digital ground) is unused here, why?

edit: here is the "digital ground and PT2399" thread: http://www.diystompboxes.com/smfforum/index.php?topic=92528.0

See my above comments on the TZF. It's worth a try, but I think you'll find that it's tricky to dial in just right. What I'm saying is breadboard it first and decide for yourself.

Connecting the anti-lockup to the digital grounds made weird noises. Also, most people in the original thread (http://www.diystompboxes.com/smfforum/index.php?topic=92515.0) didn't hear a difference.


Edit: Just came across the Archangel at circuitworkshop.com. For the record, I did come up with the idea independently and designed this without knowing about the Archangel. Great minds...

[CynicalMan]

Made a clip:

http://sites.google.com/site/distorque/home/projects/sound-clips/dimensionpdemo.mp3

The depth is set fairly high throughout this clip. You can get subtler, more "shimmery" chorus sounds, but I wanted to show how you can have the depth set high and still not get the warbling that most choruses have. At the end, I strum a chord with the dry signal disconnected from the output. You can hear the flanging sound more prominently there.

[Mark Hammer]

That's, uh, pretty dang thick.  Nice work.

[Thomeeque]

Schematic (click for bigger image)

...
The circuit definitely isn't finalized yet. I'm still tweaking it, and I'm looking forward to suggestions from you guys.

 As your LFO swings only cca 1V_PP around 4.5V and as you filter DC part from LFO signals out, you can IMO safely use 5V source as a reference voltage for U1b inverter.

Clips coming soon.

 Would you mind to make another one with significantly less depth yet (and maybe with distorted guitar too), please? I'm fan of very light (Gilmourish if you will ) use of chorus/flanger effects.. and maaaaybe (if it's not too much to ask already) with R5 = R2?

 Cheers, T.

[deadastronaut]

nice alex, really full sounding........can you get alex lifeson to come round and play red barchetta too?.....or maybe the cult?.....

this is great i just got some  pt2399's...so this is very timely....good work man...

[slacker]

That sounds great, nice work.

[DougH]

Sounds good, even on the phone speakers!

[frequencycentral]

Edit: Just came across the Archangel at circuitworkshop.com. For the record, I did come up with the idea independently and designed this without knowing about the Archangel. Great minds...

  No problem with that, yeah great minds hahaha! I assumed you'd come to it independently. They may look similar in some ways, but I think the 'intent' is what sets them apart. You were successful in your intent of a two voice chorus, I was unsuccessful in my intent of TZF. I can't remember exactly why I didn't press on with it, probably the results I got were not encouraging enough, though reading back through that post I think other stuff became priority. I'm encouraged by your work to try again someday anyway - I did draw up an Archangel 2 schematic, just never got around to trying it.

[Gurner]

It has 55 parts, compared to 48 for the Little Angel and 68 for the Small Clone. Granted, it's probably has more noise than they do, but I'm working on that.

A few extra parts could be saved if you used 5V opamps, and perhaps lobbed in a 2.5V regulator (vs resistors for 1/2 VCC ref)...... also, that would have the opamps biased at 2.5V ...which is the same as the chorus IC's pin 2 ..... (you would then likely come in at under the little angel's 48  )

[CynicalMan]

Thanks everyone! 

As your LFO swings only cca 1V_PP around 4.5V and as you filter DC part from LFO signals out, you can IMO safely use 5V source as a reference voltage for U1b inverter.

D'oh! Should have thought of that. Thanks for the tip.

Would you mind to make another one with significantly less depth yet (and maybe with distorted guitar too), please? I'm fan of very light (Gilmourish if you will ) use of chorus/flanger effects.. and maaaaybe (if it's not too much to ask already) with R5 = R2?

 Cheers, T.

Here's a clip with some subtler settings:
http://sites.google.com/site/distorque/home/projects/sound-clips/dpdemo2.mp3
Light chorus - Distorted - Gilmourish. Again, I have a chord then a lick for each setting. I'll get a sample of the TZF soon. But, as I keep telling you guys, it doesn't sound that good! 

A few extra parts could be saved if you used 5V opamps, and perhaps lobbed in a 2.5V regulator (vs resistors for 1/2 VCC ref)...... also, that would have the opamps biased at 2.5V ...which is the same as the chorus IC's pin 2 ..... (you would then likely come in at under the little angel's 48  )

I think those parts would defeat the purpose of the simplicity. Still, if we ever have another low part count FX-X...

[Gurner]

A few extra parts could be saved if you used 5V opamps, and perhaps lobbed in a 2.5V regulator (vs resistors for 1/2 VCC ref)...... also, that would have the opamps biased at 2.5V ...which is the same as the chorus IC's pin 2 ..... (you would then likely come in at under the little angel's 48  )

I think those parts would defeat the purpose of the simplicity.

I only mentioned because you specifically touched upon parts count in your opening post ....in my books soldering one 3-legged regulator (vs 4 resistors and a few blocking caps) thereby having the whole circuit biased at 2.5V (the same as the PT2399 itself)  is more simple, but hey, I'll get me coat!

[CynicalMan]

It's a good idea, but I'd prefer to stay with common parts. Also, I'm not sure if the LFO will work with a filtered Vref, it might need the resistors.

[culturejam]

Wow. Very nice project!

Add another one to the list.